Cong Hou [Mon, 2 Nov 2015 22:53:48 +0000 (22:53 +0000)]
Add a flag vectorizer-maximize-bandwidth in loop vectorizer to enable using larger vectorization factor.
To be able to maximize the bandwidth during vectorization, this patch provides a new flag vectorizer-maximize-bandwidth. When it is turned on, the vectorizer will determine the vectorization factor (VF) using the smallest instead of widest type in the loop. To avoid increasing register pressure too much, estimates of the register usage for different VFs are calculated so that we only choose a VF when its register usage doesn't exceed the number of available registers.
This is the second attempt to submit this patch. The first attempt got a test failure on ARM. This patch is updated to try to fix the failure (more specifically, by handling the case when VF=1).
Sanjay Patel [Mon, 2 Nov 2015 22:46:24 +0000 (22:46 +0000)]
[CGP] widen switch condition and case constants to target's register width
This is a follow-up from the discussion in D12965. The block-at-a-time limitation of
SelectionDAG also came up in D13297.
Without the InstCombine change from D12965, I don't expect this patch to make any
difference in the real world because InstCombine does not shrink cases like this in
visitSwitchInst(). But we need to have this CGP safety harness in place before
proceeding with any shrinkage in D12965, so we won't generate extra extends for compares.
I've opted for IR regression tests in the patch because that seems like a clearer way to
test the transform, but PowerPC CodeGen for an i16 widening test is shown below. x86
will need more work to solve: https://llvm.org/bugs/show_bug.cgi?id=22473
I'm not sure why these are happening - I added Object to the requred
libraries in tools/llvm-link/LLVMBuild.txt and the LLVM_LINK_COMPONENTS
in tools/llvm-link/CMakeLists.txt. Confirmed for my build that these
symbols come out of libLLVMObject.a. What am I missing?
Chen Li [Mon, 2 Nov 2015 22:00:15 +0000 (22:00 +0000)]
[IndVarSimplify] Rewrite loop exit values with their initial values from loop preheader
Summary:
This patch adds support to check if a loop has loop invariant conditions which lead to loop exits. If so, we know that if the exit path is taken, it is at the first loop iteration. If there is an induction variable used in that exit path whose value has not been updated, it will keep its initial value passing from loop preheader. We can therefore rewrite the exit value with
its initial value. This will help remove phis created by LCSSA and enable other optimizations like loop unswitch.
Teresa Johnson [Mon, 2 Nov 2015 21:39:10 +0000 (21:39 +0000)]
Support for ThinLTO function importing and symbol linking.
Summary:
Support for necessary linkage changes and symbol renaming during
ThinLTO function importing.
Also includes llvm-link support for manually importing functions
and associated llvm-link based tests.
Note that this does not include support for intelligently importing
metadata, which is currently imported duplicate times. That support will
be in the follow-on patch, and currently is ignored by the tests.
Cong Hou [Mon, 2 Nov 2015 21:24:00 +0000 (21:24 +0000)]
In MachineBlockPlacement, filter cold blocks off the loop chain when profile data is available.
In the current BB placement algorithm, a loop chain always contains all loop blocks. This has a drawback that cold blocks in the loop may be inserted on a hot function path, hence increasing branch cost and also reducing icache locality.
Consider a simple example shown below:
A
|
B⇆C
|
D
When B->C is quite cold, the best BB-layout should be A,B,D,C. But the current implementation produces A,C,B,D.
This patch filters those cold blocks off from the loop chain by comparing the ratio:
LoopBBFreq / LoopFreq
to 20%: if it is less than 20%, we don't include this BB to the loop chain. Here LoopFreq is the frequency of the loop when we reduce the loop into a single node. In general we have more cold blocks when the loop has few iterations. And vice versa.
James Y Knight [Mon, 2 Nov 2015 18:48:08 +0000 (18:48 +0000)]
Fix two issues in MergeConsecutiveStores:
1) PR25154. This is basically a repeat of PR18102, which was fixed in
r200201, and broken again by r234430. The latter changed which of the
store nodes was merged into from the first to the last. Thus, we now
also need to prefer merging a later store at a given address into the
target node, instead of an earlier one.
2) While investigating that, I also realized I'd introduced a bug in
r236850. There, I removed a check for alignment -- not realizing that
nothing except the alignment check was ensuring that none of the stores
were overlapping! This is a really bogus way to ensure there's no
aliased stores.
A better solution to both of these issues is likely to always use the
code added in the 'if (UseAA)' branches which rearrange the chain based
on a more principled analysis. I'll look into whether that can be used
always, but in the interest of getting things back to working, I think a
minimal change makes sense.
Silviu Baranga [Mon, 2 Nov 2015 14:41:02 +0000 (14:41 +0000)]
[SCEV][LV] Add SCEV Predicates and use them to re-implement stride versioning
Summary:
SCEV Predicates represent conditions that typically cannot be derived from
static analysis, but can be used to reduce SCEV expressions to forms which are
usable for different optimizers.
ScalarEvolution now has the rewriteUsingPredicate method which can simplify a
SCEV expression using a SCEVPredicateSet. The normal workflow of a pass using
SCEVPredicates would be to hold a SCEVPredicateSet and every time assumptions
need to be made a new SCEV Predicate would be created and added to the set.
Each time after calling getSCEV, the user will call the rewriteUsingPredicate
method.
We add two types of predicates
SCEVPredicateSet - implements a set of predicates
SCEVEqualPredicate - tests for equality between two SCEV expressions
We use the SCEVEqualPredicate to re-implement stride versioning. Every time we
version a stride, we will add a SCEVEqualPredicate to the context.
Instead of adding specific stride checks, LoopVectorize now adds a more
generic SCEV check.
We only need to add support for this in the LoopVectorizer since this is the
only pass that will do stride versioning.
This revision has introduced an issue that only affects bootstrapped compiler
when it is printing the ASM. It turns out that the new code path taken due to
legalizing a scalar_to_vector of i64 -> v2i64 exposes a missing check in a
micro optimization to change a load followed by a scalar_to_vector into a
load and splat instruction on PPC.
Craig Topper [Mon, 2 Nov 2015 07:24:37 +0000 (07:24 +0000)]
[X86] Don't pass a scale value of 0 to scatter/gather intrinsics. This causes the code emitter to throw an assertion if we try to encode it. Need to add a check to fail isel for this, but for now avoid testing it.
[PGO] Value profiling (index format) code cleanup and testing
1. Added a set of public interfaces in InstrProfRecord
class to access (read/write) value profile data.
2. Changed IndexedProfile reader and writer code to
use the newly defined interfaces and hide implementation
details.
3. Added a couple of unittests for value profiling:
- Test new interfaces to get and set value profile data
- Test value profile data merging with various scenarios.
No functional change is expected. The new interfaces will also
make it possible to change on-disk format of value prof data
to be more compact (to be submitted).
Sanjoy Das [Mon, 2 Nov 2015 02:06:01 +0000 (02:06 +0000)]
[SCEV] Fix PR25369
Have `getConstantEvolutionLoopExitValue` work correctly with multiple
entry loops.
As far as I can tell, `getConstantEvolutionLoopExitValue` never did the
right thing for multiple entry loops; and before r249712 it would
silently return an incorrect answer. r249712 changed SCEV to fail an
assert on a multiple entry loop, and this change fixes the underlying
issue.
AVX-512: Optimized SIMD truncate operations for AVX512F set.
Optimized <8 x i32> to <8 x i16>
<4 x i64> to < 4 x i32>
<16 x i16> to <16 x i8>
All these oprtrations use now AVX512F set (KNL). Before this change it was implemented with AVX2 set.
This adds support for COFF I386. This is sufficient for code execution in a
32-bit JIT, though, imported symbols need to custom lowered for the redirection.
Sanjoy Das [Sat, 31 Oct 2015 23:21:40 +0000 (23:21 +0000)]
[SCEV] Don't create SCEV expressions that break LCSSA
Prevent `createNodeFromSelectLikePHI` from creating SCEV expressions
that break LCSSA.
A better fix for the same issue is to teach SCEVExpander to not break
LCSSA by inserting PHI nodes at appropriate places. That's planned for
the future.
Rafael Espindola [Sat, 31 Oct 2015 21:44:42 +0000 (21:44 +0000)]
Don't store a Child to the first regular member.
This is a bit ugly, but has a few advantages:
* Archive is now easy to copy since there is no Archive -> Child -> Archive
loop.
* It makes it clear that we already checked for errors when finding the Child
data.
Davide Italiano [Sat, 31 Oct 2015 20:59:32 +0000 (20:59 +0000)]
[SimplifyLibCalls] Add test to ensure transform is not executed if fast-math
attribute is not present.
During my refactor in r251595 I changed the behavior of optimizeSqrt(),
skipping the transformation if the function wasn't marked with unsafe-fp-math
attribute. This fixed a bug, as confirmed by Sanjay (before the optimization
was silently executed anyway), although it wasn't my primary aim.
This commit adds a test to ensure the code doesn't break again.
Craig Topper [Sat, 31 Oct 2015 17:27:52 +0000 (17:27 +0000)]
[X86] Reduce math for index calculation for inserting and extracting subvectors and elements by exploiting the fact that all supported vector types have a power 2 number of elements.
Add a unittest for SmallDenseMap that tests assigning a SmallDenseMap when it is not small.
This complements CopyConstructorNotSmallTest. If we are testing the copy
constructor in such a way, we should also probably test assignment in the same
way.
Chandler Carruth [Fri, 30 Oct 2015 16:48:08 +0000 (16:48 +0000)]
[FunctionAttrs] Separate another chunk of the logic for functionattrs
from its pass harness by providing a lambda to query for AA results.
This allows the legacy pass to easily provide a lambda that uses the
special helpers to construct function AA results from a legacy CGSCC
pass. With the new pass manager (the next patch) the lambda just
directly wraps the intuitive query API.
Silviu Baranga [Fri, 30 Oct 2015 15:02:28 +0000 (15:02 +0000)]
[SCEV] Generalize the SCEV algorithm for creating expressions for PHI nodes
Summary:
When forming expressions for phi nodes having an incoming value from
outside the loop A and a value coming from the previous iteration B
we were forming an AddRec if:
- B was an AddRec
- the value A was equal to the value for B at iteration -1 (or equal
to the value of B shifted by one iteration, at iteration 0)
In this case, we were computing the expression to be the expression of
B, shifted by one iteration.
This changes generalizes the logic above by removing the restriction that
B needs to be an AddRec. For this we introduce two expression rewriters
that allow us to
- shift an expression by one iteration
- get the value of an expression at iteration 0
This allows us to get SCEV expressions for PHI nodes when these expressions
are not AddRecExprs.
Dehao Chen [Fri, 30 Oct 2015 05:07:15 +0000 (05:07 +0000)]
Recommit r251680 (also need to update clang test)
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
Dehao Chen [Fri, 30 Oct 2015 04:29:05 +0000 (04:29 +0000)]
Revert r251680:
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
Dehao Chen [Fri, 30 Oct 2015 02:38:29 +0000 (02:38 +0000)]
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
Alexey Samsonov [Fri, 30 Oct 2015 00:40:20 +0000 (00:40 +0000)]
Let the users of LLVMSymbolizer decide whether they want to symbolize inlined frames.
Introduce LLVMSymbolizer::symbolizeInlinedCode() instead of switching
on PrintInlining option passed to the constructor. This will be needed
once we retrun structured data (instead of std::string) from
LLVMSymbolizer and move printing logic out.
projects / llvm / log